Quantum Q&A with Ryan Anselm, a Graduating Columbia Engineer

Commencement activities have wrapped up, and another undergraduate class is taking their quantum experiences at Columbia out into the world. 

As graduating senior Ryan Anselm, SEAS’24, leaves Columbia to start his summer research with the Flatiron Institute, we asked him to reflect on what brought him to New York from Texas and to share some highlights from his time at Columbia.

What sparked your initial interest in science?

From an early age, I enjoyed reading science books my parents bought me from garage sales and book fairs. In particular, I remember reading books that explained physical phenomena like the Coriolis effect in an accessible way and pop science books that talked about Schrodinger's Cat, Maxwell's Demon, and other concepts from modern physics.

Why did you choose to study at Columbia?

I attended the University of Texas, Austin, my freshman year and then transferred. I liked reading history and philosophy books in addition to math and science, and the idea of Columbia’s core curriculum appealed to me. Columbia seemed like an ideal learning environment to be exposed to STEM and the humanities at a rigorous level simultaneously.

I was also interested in computer science. Growing up, I liked coding simple computer games, and I took some computer science classes in high school. I wanted to study math, physics, and computer science when I transferred to Columbia. It's difficult to meet all the requirements for more than one major in SEAS, so I ended up majoring in computer science with minors in applied math and applied physics.

What can you share about your undergraduate research experiences? 

Recently, I worked on compiling quantum circuits to quantum computer architectures based on reconfigurable neutral atom arrays, in which atoms with quantum bits, or qubits, encoded in them can be moved in parallel around a grid; the atoms can also have certain quantum gates applied to their qubits in parallel. In principle, arbitrary quantum circuits can be realized with a reconfigurable atom array because they have access to operations that constitute a universal set of quantum gates. 

However, in practice, qubits will tend to decohere as more operations are applied to them, so we want to use a sequence of operations that corresponds to the desired quantum circuit as efficiently as possible. To do this, we rely instead on developing heuristics that would, for example, take advantage of the ability to do certain operations in parallel.

Why is that research important?

Recent developments in neutral atom architectures have made them a plausible frontrunner candidate for building large-scale, error-corrected quantum computers. Even as the hardware improves and error rates decrease, it is still critical to the success of the architecture to have ways of efficiently translating quantum algorithms expressed in the language of quantum circuits into architecture-specific operation sequences, which can take advantage of aspects of the architecture like reconfigurability and parallel operations to reduce the number of errors accrued.

You were on a winning team at a quantum hackathon at MIT earlier this year—what was that experience like?

It was an incredibly cool experience. On Saturday morning of the hackathon, we learned what our challenge would be and spent the first several hours reading papers and figuring things out. Mid-day, our team was scheduled for a tour of an MIT lab where one of the first Bose-Einstein condensates was created. Getting to tour the lab with one of its grad students in all its mess and glory was the highlight for me. 

After that break, we continued working on our project. We ran into several issues with our code, and it was only in the early hours of the morning, around 2 or 3 AM, that things finally began to come together. We stayed up all night and were putting the finishing touches on our project up until the last minute. But by the end, our team was able to work together well and complement each other's strengths despite not knowing each other too well at the start of the event.

Our challenge was to implement and benchmark several variants of the quantum phase estimation algorithm. Essentially, we learned about several variants of a well-known quantum algorithm, implemented these variants in code, and created example problems that we could compare the performance of these algorithms on using a quantum computer simulator with realistic noise.

I understand you helped start Columbia’s Science Olympiad—why?

Science Olympiad was a very meaningful part of my high school experience because of the friendships I made and the areas of science like chemistry, geology, and astronomy that I learned from competing in it. After high school, I wanted to stay involved by volunteering at tournaments. 

Many other universities in the U.S. have student-run Science Olympiad invitationals, but Columbia didn’t, despite the number of students here who have competed in it before—it never worked out due to logistical difficulties. I'm proud that we were finally able to get it started in 2022.

What’s next for you?

I am doing more research this summer at the Center for Computational Quantum Physics at the Flatiron Institute in NYC. Then, in the fall I am starting a job as a software engineer at Salesforce. At some point down the line, I plan to return to school to pursue a PhD in quantum computing/theoretical computer science.

What has helped you take a break from academics?

I love to run! Columbia has the incredible fortune of being close to Central Park and Riverside Park, which are both extremely nice places to run that I go to often.

Any words of advice for incoming students?

Try as much as you can to choose courses by professor, not by subject, and pick a class or two to become a regular at office hours for. In my experience, office hours are an easier place than lectures to get to know other students in your class, as well as TAs and the professor.